基于直方图信息和灰度偏移网格的数字减影图像灰度矫正算法  

A Gray-Level Image Registration Algorithm for DSA Image Based on Subtraction Histogram and Gray Value Shift Mesh

作　　者：储颖[1] 唐超伦[1] 纪震[1] 牟轩沁[2] 陈思平[1] 

机构地区：[1]深圳大学信息工程学院,广东深圳518060 [2]西安交通大学图像所,陕西西安710049

出　　处：《电子学报》2006年第1期159-162,140,共5页Acta Electronica Sinica

基　　金：国家863高技术研究发展计划(No.863-306-06-06;No.2001AA114152);国家自然科学基金(No.60572100);网络与交换技术国家重点实验室开放基金;深圳市科技三项经费(No.200222)

摘　　要：本文首次提出一种利用数字减影图像直方图信息和灰度偏移网格进行灰度矫正的新的矫正算法.该算法能够有效去除灰度失真带来的蒙盈片灰度级差异,使减影图像丢失的细节信息得以重现.通过分析减影图像直方图,提取其中的灰度偏移量信息,有效消除了减影图像中的灰度级噪声.首次提出灰度偏移网格,较好地解决了数字减影图像灰度矫正过程中减影质量与矫正速度的矛盾.A new gray-level registration algorithm for digital subtraction angiography （DSA） image based on subtraction histogram （ the histogram of subtraction image） and gray value shift mesh is firstly proposed in this paper. Differences of gray values generated by gray-level distortions between mask and live images can be effectively eliminated by the presented algorithm. Therefore, the lost details in subtraction image could be restored. The gray value shift （GVS） of each pixel is calculated from the information contained in the subtraction histogram. In order to reduce the computation time,gray value shift mesh is proposed to achieve the tradeoff between registration effect and computation speed. Various experiment results have shown that this algorithm can effectively reduce artifacts and the visual quality of vessels in subtraction image is significantly improved.

关 键 词：图像配准 减影直方图 灰度矫正 数字减影血管造影术 灰度偏移网格 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]